Machine Learning Based Design of Ku Band Ridge Gap Waveguide Slot Antenna Loaded with FSS for Satellite Internet Applications

Machine learning has been used in this work for the development of a Ku band Ridge Gap Waveguide (RGW) slot antenna loaded with an FSS superstrate for satellite internet applications. The structure operates from 13.25 to 14.75 GHz with a gain beyond 10 dB using FSS superstrate loading. The developed machine learning model aims to predict the optimal length and width of the radiated slot, where both the Fractional Bandwidth (FBW) and the resonance frequency are considered objective parameters. The simulated results and the anticipated results through the machine learning algorithm are in good agreement.